Function list:
Function list:
Function list:
Function list:
accuracy = 4 corresponds to the default tuning of ABINIT. It is already a very accurate tuning.For a parallel calculation, timopt is enforced to be 0.E_min, E_med and E_max may be read from the pseudopotential file (availableonly for XML PAW atomic data files). If E_min, E_med and E_max are not givenin the pseudopotential file, ecut must be given in the input file and E_max=E_med=E_max=ecut.If the user wants to modify one of the input variable automatically tuned by accuracy,they must put it in the input file. The other input variables automatically tunedby accuracy will not be affected.accuracy = 0 means that this input variable is deactivated.
For the other values of optdriver, many of the above input variables have no meaning,so the accuracy has to be tuned by the user (e.g. for GW calculations, perform convergence studieswith respect to ecuteps and other relevant input variables).
Gives the length scales by which dimensionless primitive translations (rprim)are to be multiplied. By default, given in Bohr atomic units (1Bohr=0.5291772108 Angstroms), although Angstrom can be specified, ifpreferred, since acell has the LENGTH characteristics. See furtherdescription of acell related to the rprim input variable, thescalecart input variable, and the associated internal %rprimd input variable.
Gives the angles between directions of primitive vectors of the unit cell (indegrees), as an alternative to the input array rprim. Will be used to setup rprim, that, together with the array acell, will be used to definethe primitive vectors.
If the three angles are equal within 1.0d-12 (except if they are exactly 90degrees), the three primitive vectors are chosen so that the trigonal symmetrythat exchange them is along the z cartesian axis:
Used to define the kinetic energy cutoff which controls the number of planewaves at given k point. The allowedplane waves are those with kinetic energy lower than ecut, which translates to the following constrainton the planewave vector \vecG\vecG in reciprocal space
This is the single parameter which can have an enormous effect on the qualityof a calculation; basically the larger ecut is, the better converged thecalculation is. For fixed geometry, the total energy MUST always decrease asecut is raised because of the variational nature of the problem.
This variable activates the interpolation of the electronic eigenvalues. Itcan be used to interpolate KS eigenvalues at the end of the GS run or tointerpolate GW energies in sigma calculations (optdriver = 4). The k-pathcan be specified with kptbounds and nkpath.einterp consists of 4 entries.The first element specifies the interpolation method.
Positive values correspond to the usual choice for doing the usual ground state(GS) calculations or for structural relaxations, where the potential has to bedetermined self-consistently while negative values correspond to non-self-consistent calculations.
Controls the choice of exchange and correlation (xc). The list of XCfunctionals is given below. Positive values are for ABINIT native library ofXC functionals, while negative values are for calling the much wider set offunctionals from the ETSF LibXC library (by M. Marques), available at the LibXC home page Note that the choice made here should preferably agree with the choice made ingenerating the original pseudopotential, except for ixc = 0 (usually onlyused for debugging). A warning is issued if this is not the case.Unfortunately, pseudopotential (or PAW) generators for hybrid functionals andmGGA are currently under development, so that one usually uses GGA or LDApseudopotentials instead. The error should be limited when GGA or LDApseudopotentials with semi-core states are used.Still this is a non-controlled error. Moreover, the choices ixc = 1, 2, 3 and 7 are fits to thesame data, from Ceperley-Alder, and are rather similar, at least for spin-unpolarized systems.The choice between the non-spin-polarized and spin-polarized case is governedby the value of nsppol (see below).
NOTE: in the implementation of the spin-dependence of these functionals, andin order to avoid divergences in their derivatives, the interpolating functionbetween spin-unpolarized and fully-spin-polarized function has been slightlymodified, by including a zeta rescaled by 1.d0-1.d-6. This should affect totalenergy at the level of 1.d-6Ha, and should have an even smaller effect ondifferences of energies, or derivatives.The value ixc = 10 is used internally: gives the difference between ixc = 7 and ixc = 9, for use with an accurate RPA correlation energy.
LDA temperature-dependent functionals (do not forget to add a minus sign, as discussed above) The electronic temperature is taken from the tsmear if occopt=3 or 9 (Fermi-Dirac), from tphysel otherwise. Use of occopt=3 is recommended to have a physical meaning of the temperature. As these functionals do not provide Kxc (2nd derivative of Exc), a SCF mixing of the potential is automatically selected (iscf
Note that there is an enormous mess in the literatureconcerning the values of omega in HSE. This is due to an error in the originalpaper that stated that they had used \omega=0.15\omega=0.15. This was in fact not true,and the real value used was \omega^HF = 0.15 / \sqrt2 \sim 0.1061\omega^HF = 0.15 / \sqrt2 \sim 0.1061and \omega^PBE = 0.15 * (2.0)^1/3 \sim 0.1890\omega^PBE = 0.15 * (2.0)^1/3 \sim 0.1890.
In 2006 Krukau et al [Krukau2006] triedto clarify the situation, called HSE03 the above choice of parameters,and called HSE06 to the functional where \omega^HF=\omega^PBE\omega^HF=\omega^PBE. By testingseveral properties for atoms they reached the conclusion that the best valuefor \omega=0.11\omega=0.11. Of course, codes are just as messy as the papers. In Quantum EspressoHSE06 has the value \omega=0.106.\omega=0.106. VASP, on the other hand, uses for HSE03 thesame value \omega^HF = \omega^PBE = 0.3 (A^-1) \sim 0.1587\omega^HF = \omega^PBE = 0.3 (A^-1) \sim 0.1587,and for HSE06 \omega^HF = \omega^PBE = 0.2 (A^-1) \sim 0.1058\omega^HF = \omega^PBE = 0.2 (A^-1) \sim 0.1058.
The allowed index values are between 1 and 9999.An input variable name appended with 0 is not allowed.When ndtset == 0, this array is not used, and moreover, no input variablename appended with a digit is allowed. This array might be initialized thanksto the use of the input variable udtset. In this case, jdtset cannot be used.
b1e95dc632